As mentioned before, the microstructure of a deposit is dependent on the various deposition/processing parameters [60]. In order to satisfy the particular property needs of copper films, it is necessary to choose appropriate deposition parameters prior to carrying out electro-deposition. The electro-deposition parameters are generally considered to consist of the electrical power type and its parameters, solution composition, pH, temperature and current density.
Electrodeposition has been reported using a constant power supply at 0.6 V. The solution contained not only the conventional copper sulphate but also InCl3, GaCl3 and H2SeO3 with a pH of 2.4 at room temperature. In order to improve the adhesion between the thin film and the substrate, while decreasing the residual internal stress of the deposits, the thin film after deposition was annealed to further remove the residual stress using flowing H2S at 500 ℃ for an hour [64].
TiN film is commonly deposited as a buffer layer on the substrate prior to copper thin film electrodeposition [65]. Morisue used a constant power supply to produce copper thin films in a solution mainly containing high-grade copper sulphate. The electrolyte was de-aerated for 2 hours by bubbling through nitrogen gas in order to get high quality copper thin film without obvious voids optically [65]. Oskam demonstrated that electro-deposition using a silicon wafer substrate (n-Si 100) with a TiN buffer layer was successful in obtaining a nanocrystalline copper thin film [66]. The TiN layer was radio frequency sputter deposited at room temperature onto a silicon wafer substrate for 1 min. The aqueous solution included CuCO3·Cu(OH)2, H3BO3 and HBF4. The pH
V [66]. The experiment confirmed that TiN buffer layer can be successfully applied into copper electrodeposition in the strong acid environment, but the substrate’s pre-treatment was needed.
Some experiments were carried out using a multi-buffer layer placed between the copper thin film and substrate. A silicon wafer substrate was pre-coated with TiN and Ti, but a power supply was not used in this copper deposition. The solution composition mainly included CuCl2 (175 g), CoNO3 (6 g), HNO3 (30 ml (69%)),
ethylene diamine (2 g) and distilled water (812 g). The pH of the electroless deposition solution was 6.8. After deposition, the deposits were rinsed with deionised water and dried in an N2 stream. Two TEM samples were prepared to carry out
cross-sectional observations. One was annealed at 400 °C for 30 minutes after electrodeposition, the other was not. TEM results demonstrated that the adhesion between the deposited film and substrate was much improved by the annealing compared with the other sample which has not been annealed, as the waved-shape gap between the film and substrate was not obviously found after annealing [67].
Some of the publications do not include detailed deposition parameters. The electrolyte commonly contains Cu2+, Cl- and acid with different additives in order to achieve different properties of the deposits. Some of the experiments used copper seeds or electro-less deposition to achieve an appropriate buffer barrier prior to electro-deposition. The principle of using a seed layer is that electro-deposition onto a continuous seed layer allows the deposition of an initial layer without voids; otherwise with a discontinuous seed layer, the subsequent electro-deposited layer contained more defects and had a lamellar structure. Also poor adhesion of the seed
layer caused peeling of the thin film from the substrate surface during the planarisation process [68]. The treatment after electro-deposition varied depending on which property of the deposited copper foil was preferred. Pickling, which was used for activation prior to electro-deposition, was preferred, especially for brass substrates. As the adhesion strength was important for further mechanical testing, annealing after electro-deposition (at 300 - 400 °C lasting less than 30 minutes) [67, 68] was necessary to achieve better adhesion.
Recent research focusing on direct current electrodeposited Cu has indicated that the copper films with an average grain size ranging from 0.3-1.4 μm can be obtained with a current density ranging from 3.5-250 mA/cm2 [69, 70]。It has been recognised that the
applied current density was one of the key parameters in determining the grain size of the DC electrodeposited copper [70]. For pulsed current electrodeposited Cu, several researchers have reported that by adjusting the peak current density or the pulsed current waveform, copper films with an average grain size ranging from 0.1-0.5 μm can be obtained with a current density ranging from 5-500 mA/cm2 [70-72]. Some investigations have examined the on-time effect on microstructure [71]. A finer grain structure and stronger texture of electrodeposited copper was achieved with shorter on-time. The off-time was theoretically believed to constitute the particular stress relaxation time with recrystallization and grain growth [71].
In conclusion, the potential for copper deposition is below 1 V, and the solutions mainly consisted of copper sulphate with a pH of around 1 to 2 with some additives (Cl-). Room temperature was often used. Pre-treated activation was preferred for the substrate prior to electrodeposition and annealing was often used to decrease the
residual stress of the Cu deposits. Copper films with enhanced properties can be achieved by PC deposition rather than DC deposition. Some investigations have concerned the effect of the deposition parameters of PC prepared Cu films. Off time pf PC deposition is one of the important processing parameters, and is generally regarded as recrystallisation and grain growth time. However, the off-time effect on determining the resultant microstructure of copper films has not been particularly/fully examined yet.